Utility power meter "stuck" at 0.707 power factor

[LionelHutz]

I'm looking at a power factor correction case and the utility meter almost never reads above 0.707. There are a few measurements higher but most are either 0.707 or they are lower. The loads are all inductive yet the addition of power factor capacitors has caused the kVAR to appear higher and the power factor to appear lower compared to the metering data from before the capacitor bank addition.

The meter readings are from the meter's 15 minute fixed window demand calculations where it will give kVA, kVAR, kW and power factor for each 15 minute period. Is there anything odd the meter might be doing while calculating these demand values?

When using a power logger on site with the equipment running, the average power factor never goes below 0.9 lagging. The power factor curve recorded every second is actually quite flat varying between about 0.92 and 0.94 during the 1/2 hour I was monitoring.

Yet, utility meter data from other days with the equipment running always show a poor power factor never above 0.7 (it's not clear if this is leading or lagging).

Does anyone else believe the meter must simply be connected wrong?

If it's not a connection issue, then what could be the problem?

 

[electricpete]

3-phase or something else?
Can you narrow down to whether the error applies to kw or kva (seems like you have enough data for that comparison)

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[LionelHutz]

3-phase about 2MVA and the primary source is 69kV.

I can't directly compare the logging and the utility. The utility data was available on-line the next day but it's been off-line for the last few weeks now.

Still, it's just a bunch of motors starting full-voltage running 2 crushers. Any day with the crushers running will be about the same as any other day.

 

[desrod]

I think you have a solid case to present to the utility.

If you just added capacitors without change and measure your PF at the meter location (or close) and got a different measure, they should be the one figuring out what is wrong.

A simple calculation of your loads with typical PF should also give you a good estimate of the PF you should expect (I suppose you did that when sizing the PF correction capacitor)

We may also want to check past bills.

 

[electricpete]

One thing might be a 60 degree phase change associated with having inputs inverted and on the wrong phase.

You might have roughly a 20 degree lagging power factor angle corresdponding to pf=0.94 lagging

If you subtract 60 degrees, it looks like -40 degreess power factor angle, which would be 0.76 leading

Or not...

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

If there is are pt's and ct's, error in wye and delta connections could result in 30 degree angle error.

20 degree angle (0.94 pf) might change to 50 degree angle (0.64 pf). That doesn't quite seem to match as well.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[desrod]

At MV or HV, some utilities apply severe penalty for sudden demand increase and even if they don't last the full 15 minutes they might bill you for it.

Had a customer who managed to reduced his utility bill by properly sequencing starting up his motors in the morning.

Also remember that what you see on the utility web site might have been "processed" for billing purposes.

 

[rbulsara]

Opening a dialogue with the utility co. would seem like the thing to do. They know their meter. See what they say. Could be a wiring issue or some setting/programming on the meter. The rest is conjecture at this point.

Rafiq Bulsara

http://www.srengineersct.com

 

[rbulsara]

Also it is not very clear if you are reading at the actual utility meter and/or their on-line data?

Rafiq Bulsara

http://www.srengineersct.com

 

[rcw retired EE]

Is the meter on the high voltage side of the transformer?

If so, the meter will read the load MVAR + the tranformer MVAR losses.

 

[jghrist]

0.707 = acos(45°). Looks like the metering is using MW = Mvar all or most of the time instead of getting a true measurement.

 

[electricpete]

Of the 2 scenario's that I offered previously, the 1st is consistent with indicated power factor increasing with addition of capacitors, the 2nd is not (assuming the amount of capacitors added were based on indicated power factor). Again just a scenario for consideration.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[LionelHutz]

desrod - Interesting idea but the demand and power factor are bad all day even if they fire up the crushers and run for 3 or 4 hours.

rcwilson - true enough, there are 2 transformers metered on the high side. However, correcting the power factor on the low side should make the high side readings much better, not slightly worse.

jghrist - Yup, the meter seems to have more 0.707 power factor readings than it should, I'm just not sure why the meter would measure kW and KVAR as equal so much of the time. OK, it was doing that a lot before the caps were installed but now the power factor never gets that high.

I'm not getting paid to deal with the utility but I'm still interesting in ideas on what could be causing the problem.

 

[waross]

Are you able to see the utility meters?
Are they old electromechanical meters?
Back in the day, KVARHrs were measured with a standard KWHr meter with the addition of small phase shifting transformers.
The transformers shifted the phase of the voltage 90 degrees and made the meter respond to quadrature current rather than real current. They were a challenge to connect and the possibility of connection errors was quite high.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[tinfoil]

Some utilities 'ratchet' the demand portion of large customer's bills over an extended period. For example, one winter-peaking utility I worked for used the HIGHEST monthly peak from Dec, Jan, Feb, or the the actual last months peak (whichever is greatest) to determine the billing peak.

This peak was defined as actual kW or 90% of kVA, whichever is greater.

So you could be seeing a 'ratcheted' peak amount on your bill, rather than actual.

If you think about it, this system makes some sense, since the utility has to build enough infrastructure, from the transmission right down to the cusotmer service wires, to handle the PEAK requirements of each customer.

As others have said, ask your utility. Its free, and they usually have answers.

 

[7anoter4]

First of all, as jghrist said, you have to pick-up a "vector diagram" measuring active power, voltage and current per each phase. In the past I saw connected wrong voltage phase order.
Second: the harmonics some time could worse the pf mainly if you try to improve it by means of capacitors. The total or "true" PF accounts for the phase displacement as well as harmonic power. See:

http://ecmweb.com/mag/electric_harmonic_distortion_definitions_2/

Or:

http://books.google.com/books?id=5r...epage&q=how harmonics could reduce pf&f=false

"There is, however, a growing PF problem that cannot be fixed by adding capacitors to a circuit: switched mode power supplies which are found in personal computers and other electronic devices use rectifiers and switching transistors to regulate voltage.
These nonlinear components produce harmonics that can feed back into the circuit and, if there are a significant number of these devices, overall PF can be reduced. This is becoming a significant concern for utilities because capacitors alone will not fix the problem."
See:

http://www.pump-zone.com/motors/motors/power-factor-part-two-electricity-behaving-better.html